The physiological significance of a coronary stenosis depends on the absolute minimal cross-sectional area of the stenotic lumen. In clinical studies, the results of videodensitometric analysis of coronary stenoses have been found to correlate quite well with lesion geometry, as defined by quantitative coronary angiography, and lesion physiological significance, as defined by Doppler reactive hyperemia studies. However, more widespread application of the videodensitometric technique for analysis of coronary angiograms depends on evaluating the effects of many technical and physiological factors on the results of videodensitometric analysis. In addition, a method of defining absolute lumen area using videodensitometric techniques needs to be developed, and the technique needs to be made practical for use "on line" in the cardiac catheterization laboratory where critical decisions based on coronary anatomy are frequently made, especially during cases of angioplasty or thrombolytic therapy. In the studies proposed here, these goals will be approached using plexiglass phantoms and an animal model of coronary stenosis. Coronary stenoses will be produced by placing radiolucent plastic cylinders, with internal lumens of varying sizes (1-10 mm2) and shapes, into plexiglass phantoms or into calf coronary arteries using nonsurgical, standard cardiac catheterization techniques. Coronary cineangiograms and digital coronary angiograms of the lesions will be analyzed using videodensitometric techniques. Study protocols will be performed to evaluate the effect of the following factors on the results of videodensitometric analysis: 1) factors which influence blood/contrast mixing, 2) angulation of the vessel in the field of view, 3) lesion shape, 4) mode of acquisition of angiographic information (i.e. cineangiographic vs. digital), 5) background correction techniques, 6) degrees and methods of magnification, and 7) the phase of the cardiac cycle during which the angiogram is acquired. An attempt will then be made to define absolute lumen areas using videodensitometric techniques by use of a contrast-filled calibration catheter or by comparison of the density value of the narrowed coronary artery segment to that of a normal-appearing segment, whose size can be more easily determined using quantitative angiographic techniques.